Initialization

FINILETTI SIMONE, CADONI MATTEO, gruppo Q, Spotify Dataset¶

In [1]:
import pandas as pd;
import numpy as np;
import plotly.express as px
In [2]:
artists=pd.read_csv("data/artists.csv")
tracks = pd.read_csv("data/tracks.csv")
tracks["id_artists"] = tracks["id_artists"].apply(lambda x: x.strip('][').split(','))  #convert arrays

tracks["id_artists"]=tracks["id_artists"].apply(lambda x:x[0].replace("'",""))
combinedDf=tracks.merge(artists,left_on="id_artists",right_on="id")
combinedDf=combinedDf.rename(columns={"name_x":'title',"name_y":"artist_name", "popularity_x":"song_popularity","popularity_y":"artist_popularity","id_x":"song_id","id_y":"artist_id","macro genre":"artist_genre"})
combinedDf["tempo"]=combinedDf["tempo"].round(0)
In [ ]:
In [3]:
combinedDf.loc[combinedDf["artist_name"]=="Red Hot Chili Peppers"]
Out[3]:
song_id title song_popularity duration_ms explicit artists id_artists release_date danceability energy ... valence tempo time_signature artist_id followers genres artist_name artist_popularity TmpGenres artist_genre
218042 0ytvC9tG03obop9RXqG3af Out In L.A. - 2002 Digital Remaster 53 120667 1 ['Red Hot Chili Peppers'] 0L8ExT028jH3ddEcZwqJJ5 1984-08-10 0.772 0.888 ... 0.706 125.0 4 0L8ExT028jH3ddEcZwqJJ5 15797256.0 ['alternative rock', 'funk metal', 'funk rock'... Red Hot Chili Peppers 85 ["'alternative rock'", " 'funk metal'", " 'fun... rock
218043 0fZu9ojDt6XZLDg6dos2EM If You Want Me To Stay - Remastered 40 247333 0 ['Red Hot Chili Peppers'] 0L8ExT028jH3ddEcZwqJJ5 1985-08-16 0.829 0.791 ... 0.794 100.0 4 0L8ExT028jH3ddEcZwqJJ5 15797256.0 ['alternative rock', 'funk metal', 'funk rock'... Red Hot Chili Peppers 85 ["'alternative rock'", " 'funk metal'", " 'fun... rock
218044 3eiIQPGp6biDu91bKgAEzg Me And My Friends 53 189133 0 ['Red Hot Chili Peppers'] 0L8ExT028jH3ddEcZwqJJ5 1987 0.580 0.996 ... 0.202 100.0 4 0L8ExT028jH3ddEcZwqJJ5 15797256.0 ['alternative rock', 'funk metal', 'funk rock'... Red Hot Chili Peppers 85 ["'alternative rock'", " 'funk metal'", " 'fun... rock
218045 6NeRpRTcFftvt3j1vF4Pcw Fight Like A Brave 43 233760 0 ['Red Hot Chili Peppers'] 0L8ExT028jH3ddEcZwqJJ5 1987 0.719 0.936 ... 0.276 104.0 4 0L8ExT028jH3ddEcZwqJJ5 15797256.0 ['alternative rock', 'funk metal', 'funk rock'... Red Hot Chili Peppers 85 ["'alternative rock'", " 'funk metal'", " 'fun... rock
218046 5arVt2Wg0zbiWwAOZef2Nl Higher Ground - Remastered 67 202053 0 ['Red Hot Chili Peppers'] 0L8ExT028jH3ddEcZwqJJ5 1989-08-16 0.476 0.979 ... 0.416 140.0 4 0L8ExT028jH3ddEcZwqJJ5 15797256.0 ['alternative rock', 'funk metal', 'funk rock'... Red Hot Chili Peppers 85 ["'alternative rock'", " 'funk metal'", " 'fun... rock
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
218206 2h3DA5sidcqjywLh36RFkU Coffee Shop 41 188933 0 ['Red Hot Chili Peppers'] 0L8ExT028jH3ddEcZwqJJ5 1995-09-12 0.496 0.978 ... 0.589 121.0 4 0L8ExT028jH3ddEcZwqJJ5 15797256.0 ['alternative rock', 'funk metal', 'funk rock'... Red Hot Chili Peppers 85 ["'alternative rock'", " 'funk metal'", " 'fun... rock
218207 6jR4FKWiil1FrsfEMt7jqu Pea 39 107307 1 ['Red Hot Chili Peppers'] 0L8ExT028jH3ddEcZwqJJ5 1995-09-12 0.609 0.111 ... 0.298 177.0 4 0L8ExT028jH3ddEcZwqJJ5 15797256.0 ['alternative rock', 'funk metal', 'funk rock'... Red Hot Chili Peppers 85 ["'alternative rock'", " 'funk metal'", " 'fun... rock
218208 4VgylHaBo55VXqX4AiZiyc Deep Kick 38 393907 1 ['Red Hot Chili Peppers'] 0L8ExT028jH3ddEcZwqJJ5 1995-09-12 0.464 0.918 ... 0.322 117.0 4 0L8ExT028jH3ddEcZwqJJ5 15797256.0 ['alternative rock', 'funk metal', 'funk rock'... Red Hot Chili Peppers 85 ["'alternative rock'", " 'funk metal'", " 'fun... rock
218209 3cWEHEBIJzikXa9xPzMQkC Backwoods 35 188227 0 ['Red Hot Chili Peppers'] 0L8ExT028jH3ddEcZwqJJ5 1987 0.578 0.957 ... 0.468 111.0 4 0L8ExT028jH3ddEcZwqJJ5 15797256.0 ['alternative rock', 'funk metal', 'funk rock'... Red Hot Chili Peppers 85 ["'alternative rock'", " 'funk metal'", " 'fun... rock
218210 6ynGyjRka7Jin0xEw4Zm0k Fortune Faded 24 201267 0 ['Red Hot Chili Peppers'] 0L8ExT028jH3ddEcZwqJJ5 2003-11-18 0.524 0.946 ... 0.613 119.0 4 0L8ExT028jH3ddEcZwqJJ5 15797256.0 ['alternative rock', 'funk metal', 'funk rock'... Red Hot Chili Peppers 85 ["'alternative rock'", " 'funk metal'", " 'fun... rock

169 rows × 27 columns

Combined Analysis¶

Average value of energy per macro genre¶

In [4]:
df=combinedDf.groupby(["artist_genre"],as_index=False)["energy"].median()
fig=px.histogram(df,x="artist_genre",y="energy")
fig.update_xaxes(categoryorder="total descending")
fig.layout["xaxis"]["title"]="Generes"
fig.layout["yaxis"]["title"]="Median of energy"
fig.layout.yaxis.dtick=0.5
fig.show()

Genre and valence relationship¶

In [5]:
df=combinedDf.groupby(["artist_genre"],as_index=False)["valence"].mean()
fig=px.histogram(df,x="artist_genre",y="valence")
fig.update_xaxes(categoryorder="total descending")
fig.layout["xaxis"]["title"]="Generes"
fig.layout["yaxis"]["title"]="Valence"
fig.layout.yaxis.dtick=0.5
fig.show()

release date valence for metal music¶

In [6]:
combinedDf["release_year"]=combinedDf["release_date"].apply(lambda e:e.split("-")[0])


tmp=combinedDf.loc[(combinedDf["artist_genre"] == "metal") & (combinedDf["release_year"] >= "1971") ]

tmp["release_year"].dropna()
df=tmp.set_index(pd.to_datetime(tmp['release_year'])).resample("2Y")["valence"].mean()

tmp=combinedDf.loc[(combinedDf["artist_genre"] == "pop") & (combinedDf["release_year"] >= "1971")]

tmp["release_year"].dropna()
df2=tmp.set_index(pd.to_datetime(tmp['release_year'])).resample("2Y")["valence"].mean()
dataframe=pd.DataFrame({"valence":df})
dataframe2=pd.DataFrame({"valence":df2})
dataframe["artist_genre"]="metal"
dataframe2["artist_genre"]="pop"
dataframe=dataframe.append(dataframe2)

/tmp/ipykernel_198044/1592222919.py:17: FutureWarning:

The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
In [7]:
fig=px.line(dataframe,y="valence",color='artist_genre')

fig.layout["xaxis"]["title"]="Year"
fig.layout["yaxis"]["title"]="Valence"
fig.layout["legend"]["title"]="Genre"
fig.add_annotation(dict(font=dict(color='white',size=15),
                                        x=0,
                                        y=-0.1,
                                        showarrow=False,
                                        text="Sand, dark, negative",
                                        textangle=0,
                                        xanchor='left',
                                        xref="paper",
                                        yref="paper"))
fig.add_annotation(dict(font=dict(color='white',size=15),
                                        x=0,
                                        y=1.1,
                                        showarrow=False,
                                        text="Happy, positive",
                                        textangle=0,
                                        xanchor='left',
                                        xref="paper",
                                        yref="paper"))
fig.show()

Relationship between tempo,loudness¶

In [8]:
combinedDf["tempoType"]="Lento (0-90) bpm"
combinedDf.loc[(combinedDf["tempo"]>=90),"tempoType"]="Normal (90-130) bpm"
combinedDf.loc[(combinedDf["tempo"]>130),"tempoType"]="Speed (130-360) bpm"

df=combinedDf.groupby("tempoType",as_index=False)["valence"].mean()

fig=px.histogram(df,x="tempoType",y="valence" ,histfunc="avg",  category_orders={"Lento":"0-90",
                              "Normal": "90-130",
                              "Speed": "130-360"})


fig.layout["xaxis"]["title"]="Type of Tempo"
fig.layout["yaxis"]["title"]="Valence"


fig.show()